ARTIFICIAL REFORESTATION 23 



erage not oftener than once in five years, thus delaying 

 reforestation. Of course it will be seen that this system is 

 suitable for only such species of trees as have winged seeds. 

 The width of the strip cut should not exceed twice the 

 height of the trees unless the ground slopes rapidly away 

 from those left standing, as in that case the seeds would be 

 carried farther than if on the level or uphill. Examination 

 should be made of the trees to be left standing to ascertain 

 if they will furnish seed the year that it is proposed to cut 

 the timber. If it is not evident that seed will be forthcom- 

 ing at the right time, delay in cutting must take place, for 

 if cutting is not followed the same year by seeding, failure 

 will be very apt to result through a growth of sprouts, shrubs, 

 weeds, or from some worthless species of trees springing 

 up, and these may be so far advanced when seeding does 

 occur that germination cannot take place, or, if it does, the 

 young trees will be suppressed. To insure germination the 

 surface of the strip is sometimes gone over with some im- 

 plement that will scarify it and expose the mineral soil for 

 the seeds to fall into. However satisfactory this system 

 may prove in certain cases, it is not practiced to any great 

 extent in European countries. In Saxony, where are the 

 most perpetually productive forests in the world, in an av- 

 erage annual reforestation of sixty-nine hundred acres only 

 eight hundred are from seed sown by this method. Yet 

 there is no question but that it may serve a good purpose 

 in some situations, and especially so with such conifers as 

 are frequent and prolific seeders. On rough, rocky, or 

 swampy ground where planting young trees would be diffi- 

 cult this system may prove the best of any. Of course it is 

 obvious that in all cases where natural scattering of seeds 

 is to be depended upon, the species of trees to be grown 

 must be that of the stand from which they come, and if the 

 parent forest is of mixed species much must depend upon 

 whether all will bear seed in the same year, and, if so, 

 whether all will be scattered alike. In short, there can be 

 no satisfactory control of species likely to occur. In this 



